Spring Brassica napus canola is the most important oilseed

Transcription

1 Published August 30, 2016 RESEARCH Patterns of Heterosis in Three Distinct Inbred Populations of Spring Brassica napus Canola Habibur Rahman,* Rick A. Bennett, and Rong-Cai Yang ABSTRACT Allelic diversity of the allied species of Brassica napus L. as well as of the winter form of this species has been demonstrated to be related with increasing productivity of hybrid spring B. napus cultivars. To compare potential value of the different gene pools of Brassica species three spring B. napus inbred populations were developed by use of a B. oleracea L. line, a spring B. napus breeding line, and a winter B. napus cultivar crossed to a spring B. napus Hi- Q ; and test hybrids of these inbred lines were produced by crossing with Hi-Q as the common tester. Mid-parent heterosis (MPH) showed a negative correlation with seed yield of the inbred lines in all three populations; however, a positive correlation existed between seed yield of the inbred lines and heterosis over Hi-Q (HiQH) (or, inbred vs. hybrid yield). On average, the level of MPH in hybrid of the inbred lines derived from B. napus B. oleracea cross was twice greater than the level of heterosis found for the inbred lines derived from spring spring or winter spring B. napus crosses. The inbred population derived from winter spring cross gave highest seed yield, and this population also gave highest HiQH. The results suggested that B. oleracea and winter canola could be used in spring B. napus canola breeding for accumulating additive and non-additive effect genes for increased seed yield in hybrid cultivars. H. Rahman, Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB T6G 2P5, Canada; R.A. Bennett, Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB T6G 2P5, Canada and Agrisoma Biosciences Inc., 110 Gymnasium Place, Saskatoon, SK, S7N 0W9, Canada; R-C. Yang, Crop Research and Extension Division, Alberta Agriculture and Rural Development, Edmonton, AB T6H 5T6, and Department of Agricultural, Food and Nutritional Science, University of Alberta, Canada. Received 19 Jan Accepted 30 Mar Assigned to Associate Editor Manjit Kang. *Corresponding author (habibur.rahman@ualberta.ca). Abbreviations: HiQH, heterosis over Hi-Q; IN, lines derived from B. napus B. oleracea cross; MPH, mid-parent heterosis; NIRS, nearinfrared reflectance spectroscopy; SS, lines derived from spring spring B. napus cross; SSR, simple sequence repeat; WS, lines derived from winter spring B. napus cross. Spring Brassica napus canola is the most important oilseed crop in Canada and Australia. Genetic diversity among B. napus cultivars and lines, particularly in spring type cultivars, is known to be narrow (Hasan et al., 2006; Qian et al., 2006; Bus et al., 2011; for review see Rahman, 2013). A high level of genetic variability is critical for continued improvement of canola for seed yield and agronomic traits as well as for adaptability to changing climate. Therefore, there is a need for introduction of allelic diversity in spring B. napus canola breeding programs (for review see Rahman, 2013). Increasing genetic diversity in canola is of particular interest with regards to the breeding of hybrid cultivars. Heterosis for seed yield has been reported by different researchers in both spring (Grant and Beversdorf, 1985; Engqvist and Becker, 1991; Starmer et al., 1998) and winter (Lefort-Buson et al., 1987) type of B. napus, and genetic diversity among the inbred Published in Crop Sci. 56: (2016). doi: /cropsci Crop Science Society of America 5585 Guilford Rd., Madison, WI USA This is an open access article distributed under the CC BY-NC-ND license ( crop science, vol. 56, september october 2016

2 lines is considered essential in developing good hybrid cultivars (Diers et al., 1996; Qian et al., 2009). Among the primary and secondary gene pools of Brassica species, the winter type of B. napus has been shown to be genetically distinct from spring types (Diers and Osborn, 1994), and alleles from winter type introgressed into a spring type have been shown to have great potential for increasing seed yield in hybrid (Butruille et al., 1999; Quijada et al., 2004) and open-pollinated cultivars (Kebede et al., 2010; Rahman, 2011; Rahman and Kebede, 2012). Quijada et al. (2006) identified quantitative trait locus (QTL) alleles of winter cultivars that increase seed yield in spring B. napus test hybrids. Qian et al. (2007, 2009) suggested that some of the alleles in Chinese semi-winter type, likely introgressed from B. rapa L., might contribute to high levels of heterosis for seed yield in both spring and winter B. napus hybrids. The A and C genomes of B. napus are genetically distinct from the A genome of B. rapa and B. juncea (L.) Czern, and the C genome of B. oleracea and B. carinata A. Braun (Thormann et al., 1994; Li et al., 2004; Abel et al., 2005). Several studies have demonstrated the value of utilizing allelic diversity of the allied Brassica species for increased heterosis in B. napus (Qian et al., 2005; Li et al., 2006; Zou et al., 2010; Girke et al., 2011, 2012; Jesske et al., 2013; Li et al., 2014). For example, Zou et al. (2010) observed a high level of heterosis in test hybrids while using germplasm diversified with the A genome of B. rapa and the C genome of B. carinata. A number of researchers have also utilized genetic diversity of resynthesized B. napus (from B. rapa B. oleracea interspecific cross) for the improvement of openpollinated or hybrid winter B. napus cultivars (Gehringer et al., 2007; Radoev et al., 2008; Girke et al., 2011, 2012; Jesske et al., 2013). Udall et al. (2006) detected two QTL, where the allele from a resynthesized B. napus increased seed yield of spring B. napus. Thus, most of the studies conducted to date to evaluate different gene pools of Brassica for the improvement of spring or winter or semi-winter B. napus were focused on either the use of spring B. napus for spring type hybrid (Grant and Beversdorf, 1985; Brandle and McVetty, 1990; Engqvist and Becker, 1991; Diers et al., 1996; Cuthbert et al., 2009), or the use of winter or semi-winter B. napus for spring type hybrid (Butruille et al., 1999; Udall et al., 2004; Quijada et al., 2004, 2006; Qian et al., 2007; Kramer et al., 2009), or the use of winter or semi-winter B. napus for winter type hybrid (Lefort-Buson et al., 1987; Qian et al., 2009), or the use of the allied species for winter or semiwinter (Gehringer et al., 2007; Radoev et al., 2008; Zou et al., 2010; Fu et al., 2012; Girke et al., 2012; Li et al., 2014) or spring (Udall et al., 2004; Seyis et al., 2006) type of hybrid. However, so far, no study has been conducted to compare the value of different gene pools for increasing the level of heterosis and seed yield in spring B. napus hybrid canola. The purpose of this study was to compare three genetically distinct populations derived from spring spring and winter spring B. napus, and spring B. napus B. oleracea crosses for the level of heterosis in spring B. napus hybrids. MATERIALS AND METHODS Population Development Three inbred populations, developed from the following three crosses using B. napus (AACC genome, 2n = 38) Hi-Q as common parent, were used: Hi-Q B. oleracea var. alboglabra, Hi-Q A03-14NI, and Hi-Q Aviso. Hi-Q is a conventional non-herbicide-tolerant cultivar, whereas A03-14NI is a Clearfield herbicide-tolerant line; both are spring type, doubled haploid B. napus and developed by the Canola Program of the University of Alberta. These two parents are expected to be genetically distinct based on their breeding history. Brassica oleracea var. alboglabra (L.H. Bailey) Musil (CC genome, 2n = 18) is a spring growth-habit line; the original seeds were obtained from Lantmännen SW Seed, Sweden, from where an inbred line was developed through self-pollination of single plants for six generations. Aviso is a winter B. napus canola; seeds were obtained from Danisco Seed, Holeby, Denmark. The inbred lines derived from Hi-Q B. oleracea var. alboglabra were labeled as the IN (interspecific-derived), those from Hi-Q A03-14NI were labeled as SS (spring spring derived) and those from Hi-Q Aviso (winter spring derived) were labeled as the WS population. Because of undesired seed-quality traits (40% erucic acid content in oil; >80 µmol glucosinolate g 1 seed) of the B. oleracea parent and poor fertility in early generations of the B. napus B. oleracea interspecific hybrid progenies, selection in the segregating generations of this cross was performed for zeroerucic acid and low glucosinolate content, and for euploid B. napus plants (2n = 38). In the case of winter spring cross, plants with spring-growth habit and earliness of flowering were selected to allow for evaluation under field conditions in Canada. Aside from this, no selection for other traits was performed during the development of these three populations. Single plants were self-pollinated for four to seven generations via bag isolation in a growth chamber or greenhouse, and in this way 65 inbred lines (F 5 or F 8 ) were developed from each cross; however, randomly selected 33 lines of IN (F 8 ), 25 of SS (F 5 ), and 21 of WS (F 5 ) population were used in this study. These 33 IN, 25 SS, and 21 WS lines were genotyped, respectively, with 32, 14, and 36 polymorphic simple sequence repeat (SSR) markers, which were selected from 191 markers through screening of the parents for polymorphism. These markers were collected from various sources: Agriculture and Agri-Food Canada (AAFC) through a material transfer agreement (currently available in and the markers published by Lowe et al. (2004), Piquemal et al. (2005), Tamura et al. (2005), Iniguez-Luy et al. (2008), and Cheng et al. (2009). Genotyping of the lines was done following the method described in Bennett et al. (2012) and Kebede et al. (2010). Genotypic data showed that genetic diversity existed between and within all three populations (Supplemental Fig. S1); therefore, these 33 IN, 25 SS and 21 WS lines were used for production of test hybrids. Flow cytometric analysis of nuclear DNA content showed no significant difference between the IN lines and the B. napus parent Hi-Q (Rahman et al., 2015). crop science, vol. 56, september october

3 Test Hybrid Production Thirty-three F 8 lines of IN population, 25 F 5 lines of SS population, and 21 F 5 lines of WS population were used to produce test hybrids using Hi-Q as the common tester (female). All test hybrid seeds were produced manually through emasculation of Hi-Q, followed by cross-pollination with pollen from the IN, SS, and WS lines. Field Plot Design Test materials were grown in two replications under three environments: the Edmonton Research Station (South Campus) of the University of Alberta, Edmonton, in 2010 and 2011; and the University of Alberta St. Albert Research Farm, located approximately 25 km north of the Edmonton Research Station, in Size of each four-row plot was 1.0-m width by 2.0-m length (2.0 m 2 ), where 1 g seed was used per experimental plot. Field plots were laid out in a nested split-plot design, where main plots were the three populations (IN, SS, and WS) and subplots were inbred lines and their respective test hybrid. Inbred lines were paired with their respective test hybrids, allowing for direct inbred-hybrid comparison. Randomization of the lines/ hybrids within blocks and order of subplots was done using PROC PLAN of SAS (SAS Institute, 2010). Six plots of the parent Hi-Q were interspersed within each replication. Agronomic and Seed Quality Traits All inbred lines and their test hybrids were evaluated for the following agronomic and seed quality traits: Days to flower (d), plant height (cm), and seed yield (kg ha 1 ), and seed oil (%) and protein (%) contents. In addition, a subset of the inbred lines and their test hybrids from each population were evaluated for leaf length and width (cm), number of seeds per silique, and seed weight (g) at the South Campus site during both years. Days to flower was recorded as the number of days from seeding to when approximately 50% plants had at least one open flower. For leaf measurements, basal leaves of randomly selected 10 plants per plot were used. Likewise, 10 siliques from the middle to upperhalf of the main raceme from three plants were used to estimate the number of seeds per silique and 1000-seed weight. Seed oil and protein contents were estimated using nearinfrared spectroscopy (NIRS, FOSS NIRSystems model 6500, Foss North America, Eden Prairie, MN) on a whole-seed basis at 8.5% moisture. For this, open-pollinated bulk seed samples harvested from field plots were used. The NIRS analysis was done in the Analytical Laboratory of the Canola Program of the University of Alberta. Data Analysis Analysis of variance was done using PROC MIXED statement of SAS (SAS Institute, 2010) for individual experiments in three environments (E), each having the nested split-plot design as described above, with three population types (P) (IN, SS, and WS) being main-plot effects and two types of genetic materials (G) (inbred and test hybrid) being subplot effects. We subsequently performed a combined ANOVA to examine genotype environment interactions for all traits measured. Because unbalanced data occurred for each trait, denominator mean square and degrees of freedom were calculated using Satterwaite s synthesis method as implemented in PROC MIXED statement of SAS with the option of METHOD = Type3. The linewithin-population effect was not included in the model because we wanted to focus on the analysis at the population level. Heterosis was calculated in two ways, first being expressed as the percentage of mid-parent value (MPH, %), F 1- (P1 + P 2)/2 MPH = 100 (P + P )/2 1 2 and second being expressed as the percentage of Hi-Q (HiQH, %) F1 -HiQ HiQH = 100 HiQ The t test for significant differences between means and coefficient of correlation between two variables were calculated using EXCEL. RESULTS The ANOVA of seed yield and other agronomic (days to flower and plant height) and seed-quality traits (oil and protein) for the three types of populations (IN, WS, and SS) and two types of genetic materials (inbred and test hybrid) are presented in Table 1. Significant variation among the populations was found for all agronomic and seed quality traits; however, no differential influence of environment (nonsignificant E [environment] P [population] interaction) could be found on the populations for most of the traits. Significant differences existed among the two types of genetic materials for plant height, seed yield, and oil content; however, no differential influence of environment on these two types of genetic material (nonsignificant E G [genetic type, inbred, or hybrid] interaction) was found for most of the traits. Highly significant (P < 0.001) interaction between population and genetic material types (P G interaction) was found for all traits, suggesting that the test hybrids derived from different types of populations performed differently. The ANOVA revealed no significant E P G interaction; therefore, data of the three locations were pooled and used for estimation of MPH and HiQH. The common parent Hi-Q yielded 4609 ± 36 kg ha 1 (Table 2). Seed yields of the inbred IN and SS populations were significantly lower than that of Hi-Q (P < 0.01); however, the WS population yielded similar to Hi-Q (P > 0.05). The average yield of test hybrids was significantly (P < 0.01) higher than the inbred and mid-parent values for all three populations. The highest seed yield was found in WS inbred population, whereas the lowest seed yield occurred in IN population (Table 2). However, in the case of the hybrids, the lowest yield was observed in the case of SS population and the highest yield was still found in WS population. The difference between the hybrids of WS and SS populations was significant (P < 0.01); however, hybrid seed yield of the IN and WS populations was not crop science, vol. 56, september october 2016

4 Table 1. Analysis of variance of different traits of the inbred lines derived from Brassica napus B. oleracea interspecific cross (IN), and spring spring (SS) and winter spring (WS) crosses of B. napus and their test hybrids across three environments. Source D ays to flower, d df Error df MS F value P value Environment (E) Population (P) E P Genetic type (G) E G P G E P G Residual Plant height, cm Environment (E) Population (P) E P Genetic type (G) E G P G E P G Residual Yield, kg ha 1 Environment (E) Population (P) E P Genetic type (G) E G P G E P G Residual Seed oil, % Environment (E) Population (P) E P Genetic type (G) E G P G E P G Residual Seed protein, % Environment (E) Population (P) E P Genetic type (G) E G P G E P G Residual Because of unbalanced data set for each trait, denominator mean squares and degrees of freedom were calculated using Satterwaite s synthesis method as implemented in PROC MIXED of SAS (SAS Institute, 2010). Days to flower were measured only two environments (South Campus 2010 and South Campus 2011). significantly different, despite the fact that the IN inbred population yielded significantly lower than the WS population (P < 0.01). This inconsistency between the WS and IN populations for inbred vs. hybrid yield was at least partly attributable to greater magnitude of MPH displayed by the IN population as compared with the WS population. Average MPH in the case of IN population was almost twice greater as compared to WS and SS populations (28.7% vs. 15%) (Table 2). Highest MPH in IN population was also found in all three environments (data not shown). Among the three populations, the lowest HiQH was found for SS population (3.5%) and the highest in WS population (12.2%). The IN hybrid population exhibited twice greater magnitude of heterosis (6.6 ± 1.9%) as compared to the SS hybrid population (3.5 ± 1.7%). Variations for the level of MPH and HiQH were found in test hybrids of all three populations; the greatest variation for MPH and HiQH was found, respectively, in IN and WS hybrids and the lowest variation for both types of heterosis was found in SS hybrids. Some of the hybrids of IN and WS population yielded more than 25% of the spring canola parent Hi-Q. A negative relationship between MPH and seed yield of the inbred lines was found in all three populations; however, though weak, a positive relation was found between HiQH and inbred yield in all three populations (Fig. 1). Indeed, the relationship between inbred yield and HiQH also exactly reflects the relationship between inbred and hybrid seed yield (Supplemental Fig. S2). Correlation between genetic distance of the inbred lines from Hi-Q and MPH (r = to 0.034) or HiQH (r = to 0.007) was weak in all three populations as well as in pooled data of the three populations (r = for MPH and for HiQH) (Supplemental Fig. S3); however, some of the lines with intermediate genetic distance tended to display greater heterosis than the lines with low or high genetic distance from Hi-Q. This result, however, needs to be treated with caution as the estimation of genetic distance was based on only a limited number of polymorphic markers available in this study. Among the three inbred populations, the IN population flowered earliest, whereas the WS population flowered latest; the same scenario was also observed in case of the three hybrid populations. The test hybrid populations, on average, flowered slightly earlier than the average flowering time of the two parents; however, none of the hybrids flowered earlier than the earliest parent. The positive HiQH in WS population was attributable to the fact that the WS inbred population flowered later than Hi-Q as opposed to the other two populations, which flowered earlier than Hi-Q. For the plant traits, about 5 to 6% MPH was found for plant height in all three populations. For HiQH, the SS and WS populations exhibited about 4 to 5% heterosis; however, slightly negative HiQH ( 3.0 ± 0.8%) was found crop science, vol. 56, september october

5 Table 2. Agronomic traits (mean ± SE) of the parents and hybrids of Brassica napus populations derived from spring B. napus B. oleracea (IN), spring B. napus canola spring B. napus canola (SS), and winter B. napus canola spring B. napus canola (WS) crosses, and heterosis over mid-parent (MHP, %) and over Hi-Q (HiQH, %) for these traits. Population and statistic Days to flower Plant height Yield Leaf length Leaf width No. seeds/silique d cm kg ha 1 c m Hi-Q 55.8 ± ± ± ± ± ± 0.2 IN: Lines 50.7 ± ± ± ± ± ± 1.7 IN: MP 53.1 ± ± ± ± ± ± 0.8 IN: Hybrid 53.1 ± ± ± ± ± ± 1.1 t test **/NS **/** **/** **/** **/** NS/NS IN: MPH, % 0.1 ± ± ± ± ± ± 4.6 IN: HiQH, % 4.8 ± ± ± ± ± ± 3.4 t test ** ** ** ** * NS SS: Lines 53.7 ± ± ± ± ± ± 0.6 SS: MP 54.7 ± ± ± ± ± ± 0.3 SS: Hybrid 54.4 ± ± ± ± ± ± 0.5 t test NS/NS **/** **/** */* */NS **/NS SS: MPH, % 0.6 ± ± ± ± ± ± 0.8 SS: HiQH, % 2.4 ± ± ± ± ± ± 1.5 t test ** ** ** NS NS ** WS: Lines 58.2 ± ± ± ± ± ± 1.1 WS: MP 57.0 ± ± ± ± ± ± 0.5 WS: Hybrid 56.3 ± ± ± ± ± ± 0.9 t test **/* */** **/** NS/** NS/* */NS WS: MPH, % 1.2 ± ± ± ± ± ± 2.6 WS: HiQH, % 1.0 ± ± ± ± ± ± 2.8 t test ** NS NS * NS NS * Significant at P < ** Significant at P < Hi-Q = common spring B. napus canola parent; IN = lines derived from B. napus B. oleracea interspecific cross; SS = lines derived from spring B. napus canola spring B. napus canola cross; WS = lines derived from winter B. napus canola spring B. napus canola cross. MP = mid-parent value; MPH (%) = percent heterosis over midparent; HiQH (%) = percent heterosis over Hi-Q. Population size: IN: n = 33 for days to flower, plant height and seed yield; n = 23 for leaf length and leaf width; n = 8 for number seeds/silique and 1000-seed weight. SS: n = 25 for days to flower, plant height and seed yield; n = 16 for leaf length and leaf width; n = 8 for number seeds/silique and 1000-seed weight. WS: n = 21 for days to flower, plant height and seed yield; n = 15 for leaf length and leaf width; n = 8 for number seeds/silique and 1000-seed weight. t test: Line vs. hybrid or MP vs. hybrid. NS: not significant. in the IN hybrid population. Hybrids of IN population were significantly shorter (P < 0.01) than the hybrids of SS and WS populations; however, no significant difference between the hybrids of SS and WS population was found. In the case of leaf size, about 5 to 8% MPH was found in all three populations. The IN inbred population had smallest leaf size and displayed almost no HiQH. On the other hand, the WS inbred population had the largest leaf size and also displayed the largest HiQH (>10%). For the seed traits, no significant difference between test hybrid and inbred populations was found for seed weight; the difference between MPH and HiQH for this trait was also nonsignificant (Table 3). Among the three populations, the SS inbred population had highest oil content but lowest protein content, whereas the IN population had lowest oil content and highest protein content this well-known negative correlation between these two traits is also evident in these populations. Sum of oil and protein contents was highest in SS population. The hybrid populations did not differ significantly from the mid-parent value for both oil and protein contents in all three cases. This was also evident from almost no MPH having been observed for these seed traits. DISCUSSION The experimental design of this study was such that spring B. napus Hi-Q was used as a common parent for the development of all three inbred populations (IN, WS, and SS), as well as common parent for the development of the three test-hybrid populations. This design allowed us to evaluate the relative importance of the different gene pools for MPH and hybrid yield (in this case HiQH). Based on this design, any difference observed between the three hybrid populations for heterosis or hybrid yield is expected to be because of allelic variation in the inbred lines originating from B. oleracea or spring B. napus or winter B. napus. In this study, seed yield of the IN inbred population was significantly lower than that of the WS population; crop science, vol. 56, september october 2016

6 Fig. 1. Relationship between inbred lines, derived from Brassica napus B. oleracea interspecific cross (IN), and spring spring (SS) and winter spring (WS) crosses of B. napus, and mid-parent heterosis (MPH) and heterosis over common B. napus parent Hi-Q (HiQH) for seed yield. Relationship between seed yield of the inbred lines and their hybrids (Supplemental Fig. S2) is exactly same as the relationship between inbred lines and HiQH. crop science, vol. 56, september october

7 Table 3. Seed traits (mean ± SE) of the parents and hybrids of Brassica napus populations derived from spring B. napus B. oleracea (IN), spring B. napus canola spring B. napus canola (SS), and winter B. napus canola spring B. napus canola (WS) crosses, and heterosis over mid-parent (MHP, %) and over Hi-Q (HiQH, %) for these traits. Population 1000-seed and statistic weight Oil Protein Oil + Protein g % Hi-Q 4.30 ± ± ± ± 0.01 IN: Lines 4.97 ± ± ± ± 0.3 IN: MP 4.63 ± ± ± ± 0.2 IN: Hybrid 4.69 ± ± ± ± 0.2 t test NS/NS **/NS */NS NS/NS IN: MPH, % 1.37 ± ± ± ± 0.1 IN: HiQH, % 9.06 ± ± ± ± 0.3 t test NS ** ** * SS: Lines 4.46 ± ± ± ± 0.1 SS: MP 4.38 ± ± ± ± 0.1 SS: Hybrid 4.33 ± ± ± ± 0.1 t test NS/NS NS/NS **/NS **/** SS: MPH, % 1.18 ± ± ± ± 0.1 SS: HiQH, % 0.58 ± ± ± ± 0.1 t test NS ** ** NS WS: Lines 4.73 ± ± ± ± 0.2 WS: MP 4.52 ± ± ± ± 0.1 WS: Hybrid 4.61 ± ± ± ± 0.2 t test NS/NS NS/NS NS/NS */NS WS: MPH, % 2.35 ± ± ± ± 0.2 WS: HiQH, % 7.17 ± ± ± ± 0.3 t test NS NS * ** * Significant at P < ** Significant at P < Hi-Q = common spring B. napus canola parent; IN = lines derived from B. napus B. oleracea interspecific cross; SS = lines derived from spring B. napus canola spring B. napus canola cross; WS = lines derived from winter B. napus canola spring B. napus canola cross. MP = mid-parent value; MPH (%) = percent heterosis over mid-parent; HiQH (%) = percent heterosis over Hi-Q. Population size: IN, n = 33; SS, n = 25; and WS, n = 21. t test: Line vs. hybrid or MP vs. hybrid. NS: not significant. however, the difference between hybrid seed yield of these two populations was not significant. This nonsignificant difference largely resulted from non-additive effect of the genes contributing to heterosis, as evident from significant difference in the level of MPH found among these two populations. The occurrence of lower levels of MPH in WS and SS populations as compared to the IN population might be due to the depletion of some of the alleles (e.g., recessive), which are capable of exhibiting non-additive effect (e.g., overdominance) for heterosis, from the highyielding inbred line cultivars (e.g., Hi-Q, Aviso, and A03-14NI) due to stringent selection over cycles of breeding for the alleles (e.g., dominant) which are favorable for high seed yield in line (homozygous) cultivars. However, this type of alleles (unfavorable for line cultivars but favorable for heterosis) can be found in IN population, that is, in B. oleracea, more frequently as compared to winter and spring B. napus canola; this is evident from greater MPH observed in IN hybrids as compared to WS and SS hybrids, and the negative correlation between inbred seed yield and MPH (Fig. 1). This is also evident from a high level of polymorphisms observed in the IN cross (Supplemental Table S1). Of a total of 70 SSR markers assayed across three crosses, 46% (32/70), 51% (36/70), and 20% (14/70) of the markers were polymorphic in IN, WS, and SS crosses, respectively. Introduction of this type of allele in hybrid parent lines can increase the level of heterosis in hybrid cultivars. Riaz et al. (2001) evaluated 12 hybrids involving parents from different genetic diversity groups and reported highest MPH and high-parent heterosis for the hybrids that involved a parental line derived from a B. napus B. oleracea cross that was developed by Quazi (1988). Recently, Li et al. (2014) also demonstrated the value of the B. oleracea gene pool for use in breeding of Chinese semi-winter B. napus hybrid cultivars. By use of a doubled-haploid (DH) population derived from a cross between a European winter canola and a resynthesized B. napus line (resynthesized from B. oleracea var. sabellica and B. rapa ssp. pekinensis), Radoev et al. (2008) found mainly dominance and overdominance effects of the genes contributing to heterosis. The low seed yield in the IN inbred population was apparently attributable to introduction of several unfavorable alleles from B. oleracea, and this is not very uncommon for the inbred lines derived from interspecific crosses. For example, Girke et al. (2012) found that resynthesized B. napus lines, on average, showed poor performance for seed yield; however, this type of materials can show good potential for high seed yield in hybrids. Among the three populations, the WS inbred population and its hybrid population yielded higher than the other two populations. This agrees with the results of Butruille et al. (1999), Quijada et al. (2004), Kebede et al. (2010), and Rahman and Kebede (2012), which indicated that the European winter canola could be used for broadening of genetic diversity in spring canola as well as for increasing seed yield of spring canola hybrids and open-pollinated line cultivars. The use of winter canola in breeding of spring hybrid cultivar can be advantageous as no stringent selection for zero erucic acid in oil and low glucosinolate content in seed meal (Kebede et al., 2010; Rahman, 2011; Rahman and Kebede, 2012) is needed as well as no difficulty to be encountered for the development of euploid (2n = 38) B. napus inbred lines from this type of crosses. This is in contrast to the use of allied species in breeding of hybrid-parents which encounter much difficulty for the development of a canola quality euploid line (Bennett et al., 2012; for review see Rahman, 2013; Rahman et al., 2015). Breeding in the past has always focused on selection for favorable (e.g., dominant) alleles for the development of winter and spring B. napus canola line cultivars and this might have eroded their contrasting crop science, vol. 56, september october 2016

8 (recessive) alleles from the breeding populations, which can potentially exhibit non-additive effect and contribute to heterosis. This might be the reason for the lack of strong positive correlation between inbred and hybrid yield; however, the observed positive correlation between inbred and hybrid yield in all three populations confirmed the importance of general combining ability of the parents for increasing seed yield in hybrid cultivars as suggested by Diers et al. (1996) and Qian et al. (2007, 2009). Among the three populations, strongest positive correlation was found in WS population, suggesting that general combining ability of the inbred lines can be improved by the use of European winter canola in the breeding of spring canola. A decrease in genetic diversity across cycles of breeding has also been reported in other crops, such as maize, where the use of exotic germplasm in hybrid breeding is well documented (for review see Goodman, 1999; Reif et al., 2010). Days to flower in spring type B. napus is a quantitative trait controlled by genes with additive, dominance, and epistatic effects (Ringdahl et al., 1986; Long et al., 2007); this trait correlates well with days to maturity in both B. napus (Miller 2001) and B. juncea (Mahmood et al., 2007). Involvement of genes with dominance effect relative to earliness of flowering is evident from average negative MPH in all three populations as well as negative HiQH in IN and SS populations. This type of gene effect is especially important for those areas of North America where earliness of flowering and maturity are an important breeding objective for the development of hybrid canola cultivars. Long et al. (2007) also found that 10% of the total genetic effect for flowering time was contributed by dominance genes in winter/semi-winter B. napus. The hybrids of all three populations also displayed an increase in plant height and leaf size. This observation was similar to reports of vigorous growth of European spring Canadian spring B. napus hybrids compared with their parents as noted by Cuthbert et al. (2009). Heterosis for biomass has also been reported in European winter B. rapa (Ofori et al., 2012), vegetable B. rapa (Dong et al., 2007), as well as in hybrids of B. napus B. rapa cross (Liu et al., 2002). The slight negative heterosis for plant height observed in case of IN population indicated that B. oleracea might carry alleles for reducing plant height in hybrid cultivars. In the case of the seed traits, no apparent difference between hybrid and mid-parent values was found for seed oil and protein contents, which was primarily for the reason that these two traits were largely controlled by genes with additive effect (for review, see Rahman et al., 2013). One of the limitations of this study was the size of the field plots. Normally, at minimum, 5 m 2 plots are used for testing of spring canola inbred lines and test hybrids produced by use of a male-sterility system (Quijada et al., 2006; Udall et al., 2006). The hypothesis laid out in this research necessitated production of test hybrids of all combinations through manual emasculation of the female, followed by hand pollination with pollen from the male lines. Manual production of seeds of a large number of test hybrids is a tedious and resource-demanding task and, therefore, only limited quantity of seeds of each test hybrid could be produced for field experimentation. Also, the main focus of our study was on population level and individual lines in a population served as a replication. Cuthbert et al. (2009) has demonstrated that valuable information can also be obtained from trialling of spring B. napus test hybrids in single-row 3-m long plots (0.5 g seed per plot). CONCLUSIONS In conclusion, of the three gene pools investigated in this study, the B. oleracea and the winter canola showed good potential for increasing seed yield in hybrid cultivars. Alleles exerting non-additive effect (e.g., overdominance) in the genetic control of heterosis were found in B. oleracea, whereas the alleles for increased general combining ability of the hybrid parent lines were found in winter canola to a greater extent than the other populations. The heterotic effect of the IN and WS alleles could possibly be even greater, should alternative tester lines, unrelated to Hi-Q, be used to generate the test hybrids. Thus, the continued improvement of spring B. napus hybrid cultivars will likely include the utilization of winter types and the allied species, including the C-genome of B. oleracea in breeding. Supplemental Material Available Supplemental information is available with the online version of this manuscript. Acknowledgments Habibur Rahman gratefully acknowledges Natural Sciences and Engineering Research Council of Canada (NSERC) for funding this project, and Canada Foundation for Innovation for support for infrastructure development in the Canola Program of the University of Alberta. Authors are thankful to An Vo, Salvador Lopez, Berisso Kebede, and other staff from the Canola Program for assistance in various routine works. References Abel, S., C. Möllers, and H.C. Hecker Development of synthetic Brassica napus lines for the analysis of fixed heterosis in allopolyploid plants. Euphytica 146: doi: / s Bennett, R.A., G. Séguin-Swartz, and H. Rahman Broadening genetic diversity in canola: Towards the development of canola quality Brassica oleracea. Crop Sci. 52: doi: /cropsci Brandle, J.E., and P.B.E. McVetty Geographical diversity, parental selection and heterosis in oilseed rape. Can. J. Plant Sci. 70: doi: /cjps crop science, vol. 56, september october

9 Bus, A., N. Körber, R.J. Snowdon, and B. Stich Patterns of molecular variation in a species-wide germplasm set of Brassica napus. Theor. Appl. Genet. 123: doi: /s Butruille, D.V., R.P. Guries, and T.C. Osborn Increasing yield of spring oilseed rape hybrids through introgression of winter germplasm. Crop Sci. 39: doi: / cropsci x Cheng, X., J. Xu, S. Xia, J. Gu, Y. Yang, J. Fu et al Development and genetic mapping of microsatellite markers from genome survey sequences in Brassica napus. Theor. Appl. Genet. 118: doi: /s Cuthbert, R.D., G. Crow, and P.B.E. McVetty Assessment of agronomic performance and heterosis for agronomic traits in hybrid high erucic acid rapeseed (HEAR). Can. J. Plant Sci. 89: doi: /cjps08160 Diers, B.W., P.B.E. McVetty, and T.C. Osborn Relationship between heterosis and genetic distance based on restriction fragment length polymorphism markers in oilseed rape (Brassica napus L.). Crop Sci. 36: doi: /cropsci X x Diers, B.W., and T.C. Osborn Genetic diversity of oilseed Brassica napus germ plasm based on restriction fragment length polymorphisms. Theor. Appl. Genet. 88: doi: / BF Dong, D.-K., J.-S. Cao, K. Shi, and L.-C. Liu Overdominance and epistasis are important for the genetic basis of heterosis in Brassica rapa. HortScience 42: Engqvist, G.M., and H.C. Becker Heterosis and epistasis in rapeseed estimated from generation means. Euphytica 58: doi: /bf Fu, D., W. Qian, J. Zou, and J. Meng Genetic dissection of intersubgenomic heterosis in Brassica napus carrying genomic components of B. rapa. Euphytica 184: doi: / s Gehringer, A., R. Snowdon, T. Spiller, P. Basunanda, and W. Friedt New oilseed rape (Brassica napus) hybrids with high levels of heterosis for seed yield under nutrient-poor conditions. Breed. Sci. 57: doi: /jsbbs Girke, A., A. Schierholt, and H.C. Becker Extending the rapeseed genepool with resynthesized Brassica napus L. I: Genetic diversity. Genet. Resour. Crop Evol. 59: doi: / s Girke, A., A. Schierholt, and H.C. Becker Extending the rapeseed gene pool with resynthesized Brassica napus II. Heterosis. Theor. Appl. Genet. 124: doi: /s Goodman, M.M Broadening the genetic diversity in maize breeding by use of exotic germplasm. In: J.G. Coors and S. Pandey, editors, The genetics and exploitation of heterosis in crops. ASA, CSSA, and SSSA, Madison, WI. p Grant, I., and W.D. Beversdorf Heterosis and combining ability estimates in spring-planted oilseed rape (Brassica napus L.). Can. J. Genet. Cytol. 27: doi: /g Hasan, M.F., F. Seyis, A.G. Badani, J. Pons-Kühnemann, W. Friedt, W. Lühs, and R.J. Snowdon Surveying genetic diversity in the Brassica napus L. gene pool using SSR markers. Genet. Resour. Crop Evol. 53: doi: /s Iniguez-Luy, F.L., A.V. Voort, and T.C. Osborn Development of a set of SSR markers derived from genomic sequence of a rapid cycling Brassica oleracea L. genotype. Theor. Appl. Genet. 117: doi: /s Jesske, T., B. Olberg, A. Schierholt, and H.C. Becker Resynthesized lines from domesticated and wild Brassica taxa and their hybrids with B. napus L.: Genetic diversity and hybrid yield. Theor. Appl. Genet. 126: doi: /s y Kebede, B., M.R. Thiagarajah, C. Zimmerli, and M.H. Rahman Improvement of open-pollinated spring rapeseed (Brassica napus L.) through introgression of genetic diversity from winter rapeseed. Crop Sci. 50: doi: /cropsci Kramer, C.C., H. Polewicz, and T.C. Osborn Evaluation of QTL alleles from exotic sources for hybrid seed yield in the original and different genetic backgrounds of spring-type Brassica napus L. Mol. Breed. 24: doi: /s x Lefort-Buson, M.B., B. Guillot-Lemoine, and Y. Dattée Heterosis and genetic distance in rapeseed (Brassica napus L.): Crosses between European and Asiatic selfed lines. Genome 29: doi: /g Li, M., X. Chen, and J. Meng Intersubgenomic heterosis in rapeseed production with a partial new-typed Brassica napus containing subgenome A r from B. rapa and C c from Brassica carinata. Crop Sci. 46: doi: /cropsci Li, M., W. Qian, J. Meng, and Z. Li Construction of novel Brassica napus genotypes through chromosomal substitution and elimination using interploid species hybridization. Chromosome Res. 12: doi: /b:chro Li, Q., Q. Zhou, J. Mei, Y. Zhang, J. Li, Z. Li et al Improvement of Brassica napus via interspecific hybridization between B. napus and B. oleracea. Mol. Breed. 34: doi: / s Liu, R., W. Qian, and J. Meng Association of RFLP markers and biomass heterosis in trigenomic hybrids of oilseed rape (Brassica napus B. campestris). Theor. Appl. Genet. 105: doi: /s x Long, Y., J. Shi, D. Qiu, R. Li, C. Zhang, J. Wang et al Flowering time quantitative trait loci analysis of oilseed Brassica in multiple environments and genomewide alignment with Arabidopsis. Genetics 177: Lowe, A.J., C.L. Moule, M. Trick, and K.J. Edwards Efficient large-scale development of microsatellites for marker and mapping applications in Brassica crop species. Theor. Appl. Genet. 108: doi: /s Mahmood, T., M.H. Rahman, G. Stringam, F. Yeh, and A. Good Quantitative trait loci for early maturity and their potential in breeding for earliness in Brassica juncea. Euphytica 154: doi: /s Miller, T.A Agronomic and quality performance of three doubled haploid lines derived from a Brassica napus/brassica rapa interspecific cross. M.Sc. thesis. Univ. of Alberta, Edmonton. Ofori, A., A. Schierholt, and H.C. Becker Biomass yield and heterosis of crosses within and between European winter cultivars of turnip rape (Brassica rapa L.). J. Appl. Genet. 53: doi: /s Piquemal, J., E. Cinquin, F. Couton, C. Rondeau, E. Seignoret, I. Doucet et al Construction of an oilseed rape (Brassica napus L.) genetic map with SSR markers. Theor. Appl. Genet. 111: doi: /s Qian, W., X. Chen, D. Fu, J. Zou, and J. Meng Intersubgenomic heterosis in seed yield potential observed in a new type of Brassica napus introgressed with partial Brassica rapa genome. Theor. Appl. Genet. 110: doi: /s crop science, vol. 56, september october 2016

10 Qian, W., Q. Li, J. Noack, O. Sass, J. Meng, M. Frauen, and C. Jung Heterotic patterns in rapeseed (Brassica napus L.): II. Crosses between European winter and Chinese semi-winter lines. Plant Breed. 128: doi: /j x Qian, W., J. Meng, M. Li, M. Frauen, O. Sass, J. Noack, and C. Jung Introgression of genomic components from Chinese Brassica rapa contributes to widening the genetic diversity in rapeseed (B. napus L.), with emphasis on the evolution of Chinese rapeseed. Theor. Appl. Genet. 113: doi: /s Qian, W., O. Sass, J. Meng, M. Li, M. Frauen, and C. Jung Heterotic patterns in rapeseed (Brassica napus L.): I. Crosses between spring and Chinese semi-winter lines. Theor. Appl. Genet. 115: doi: /s x Quazi, M.H Interspecific hybrids between Brassica napus L. and B. oleracea L. developed by embryo culture. Theor. Appl. Genet. 75: doi: /bf Quijada, P.A., J.A. Udall, B. Lambert, and T.C. Osborn Quantitative trait analysis of seed yield and other complex traits in hybrid spring rapeseed (Brassica napus L.): 1. Identification of genomic regions from winter germplasm. Theor. Appl. Genet. 113: doi: /s Quijada, P.A., J.A. Udall, H. Polewicz, R. Vogelzang, and T.C. Osborn Phenotypic effects of introgressing French winter germplasm into hybrid spring canola. Crop Sci. 44: doi: /cropsci Radoev, M., H.C. Becker, and W. Ecke Genetic analysis of heterosis for yield and yield components in rapeseed (Brassica napus L.) by quantitative trait locus mapping. Genetics 179: doi: /genetics Rahman, H Use of European winter canola Brassica napus L. for the improvement of Canadian spring canola: A practical breeding example. In: Proceedings of the 13th Inter national Rapeseed Congress, Prague. 5 9 June Int. Consultative Group for Res. on Rapeseed, Paris. p Rahman, H Review: Breeding spring canola (Brassica napus L.) by the use of exotic germplasm. Can. J. Plant Sci. 93: doi: /cjps Rahman, H., R.A. Bennett, and G. Se guin-swartz Broadening genetic diversity in Brassica napus canola: Development of canola quality B. napus lines from B. napus B. oleracea var. alboglabra interspecific cross. Can. J. Plant Sci. 95: doi: / cjps Rahman, H., J. Harwood, and R. Weselake Increasing seed oil content in Brassica species through breeding and biotechnology. Lipid Technology 25: doi: /lite Rahman, H., and B. Kebede Improvement of spring canola Brassica napus (L.) by use of winter canola. J. Oilseed Brassica 3:1 17. Reif, J.C., S. Fischer, T.A. Schrag, K.R. Lamkey, D. Klein, B.S. Dhillon et al Broadening the genetic base of European maize heterotic pools with US Cornbelt germplasm using field and molecular marker data. Theor. Appl. Genet. 120: doi: /s Riaz, A., G. Li, Z. Quresh, M.S. Swati, and C.F. Quiros Genetic diversity of oilseed Brassica napus inbred lines based on sequence-related amplified polymorphism and its relation to hybrid performance. Plant Breed. 120: doi: / j x Ringdahl, E.A., P.B.E. McVetty, and J.L. Sernyk Inheritance of earliness, height, and leaf number in crosses of early maturing rapeseed. Can. J. Genet. Cytol. 28: doi: / g SAS Institute SAS Statistical Analysis Software Version SAS Inst., Cary, NC. Seyis, F., W. Friedt, and W. Lühs Yield of Brassica napus L. hybrids developed using resynthesized rapeseed material sown at different locations. Field Crops Res. 96: doi: /j. fcr Starmer, K.P., J. Brown, and J.B. Davis Heterosis in spring canola hybrids grown in Northern Idaho. Crop Sci. 38: doi: /cropsci x x Tamura, K., M. Nishioka, M. Hayashi, Z. Zhang, C. Lian, T. Hougetsu, and K. Harada Development of microsatellite markers by ISSR-suppression-PCR method in Brassica rapa. Breed. Sci. 55: doi: /jsbbs Thormann, C.E., M.E. Ferreira, L.E.A. Camargo, J.G. Tivang, and T.C. Osborn Comparison of RFLP and RAPD markers to estimating genetic relationships within and among cruciferous species. Theor. Appl. Genet. 88: doi: / BF Udall, J.A., P.A. Quijada, B. Lambert, and T.C. Osborn Quantitative trait analysis of seed yield and other complex traits in hybrid spring rapeseed (Brassica napus L.): 2. Identification of alleles from unadapted germplasm. Theor. Appl. Genet. 113: doi: /s Udall, J.A., P.A. Quijada, H. Polewicz, R. Vogelzang, and T.C. Osborn Phenotypic effects of introgressing Chinese winter and resynthesized Brassica napus L. germplasm into hybrid spring canola. Crop Sci. 44: doi: /cropsci Zou, J., J. Zhu, S. Huang, E. Tian, Y. Xiao, D. Fu et al Broadening the avenue of intersubgenomic heterosis in oilseed Brassica. Theor. Appl. Genet. 120: doi: /s crop science, vol. 56, september october